Lagrangian flow trajectories are a ubiquitous tool in atmospheric and oceanographic science. These trajectories are used to analyse dynamical and mixing processes, but also to study e.g. chemistry and construct detailed case studies for Large-Eddy Simulation. However, a Lagrangian perspective can also be useful for solving the equations of motion. In this talk, I will discuss collaborative work to go beyond the semi-Lagrangian approach (which uses a Lagrangian framework for advection, but is still largely grid-based) and build a geophysical model based entirely around Lagrangian parcels.

The parcels in this Moist Parcel-In-Cell (MPIC) model represent both the thermodynamic and the dynamical prognostic properties of the flow. An efficient grid-based solver calculates parcel advection velocities. The Lagrangian approach of MPIC has a number of advantages: thermodynamic/tracer properties and their correlations are naturally conserved, and the amount of mixing between parcels can be explicitly controlled. I will present results demonstrating the performance of MPIC relative to the Met Office/NERC cloud model for an individual convective cloud, and talk about ongoing work to address the challenges in applying this method to more realistic case studies.

* Work with David Dritschel (University of St Andrews), Alan Blyth (University of Leeds), Doug Parker (University of Leeds), Matthias Frey (University of St Andrews) and Gordon Gibb (EPCC).